CN114237226B - Navigation escaping method under complex environment area coverage of group robot - Google Patents
Navigation escaping method under complex environment area coverage of group robot Download PDFInfo
- Publication number
- CN114237226B CN114237226B CN202111384047.9A CN202111384047A CN114237226B CN 114237226 B CN114237226 B CN 114237226B CN 202111384047 A CN202111384047 A CN 202111384047A CN 114237226 B CN114237226 B CN 114237226B
- Authority
- CN
- China
- Prior art keywords
- robot
- grid
- navigation
- coverage
- value
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Active
Links
- 238000000034 method Methods 0.000 title claims abstract description 28
- 238000004422 calculation algorithm Methods 0.000 claims abstract description 22
- 230000001413 cellular effect Effects 0.000 claims abstract description 9
- 230000000694 effects Effects 0.000 claims description 39
- 210000002569 neuron Anatomy 0.000 claims description 33
- 210000004027 cell Anatomy 0.000 claims description 32
- 230000000638 stimulation Effects 0.000 claims description 7
- 230000004888 barrier function Effects 0.000 claims description 5
- 238000013528 artificial neural network Methods 0.000 claims description 4
- 238000003062 neural network model Methods 0.000 claims description 4
- 230000007246 mechanism Effects 0.000 abstract description 7
- 230000008569 process Effects 0.000 abstract description 7
- 238000004891 communication Methods 0.000 abstract description 5
- 238000004364 calculation method Methods 0.000 abstract description 4
- 230000002457 bidirectional effect Effects 0.000 abstract description 3
- 238000004088 simulation Methods 0.000 description 10
- 238000001514 detection method Methods 0.000 description 4
- 238000010586 diagram Methods 0.000 description 4
- 238000005070 sampling Methods 0.000 description 4
- 230000008901 benefit Effects 0.000 description 2
- 230000005284 excitation Effects 0.000 description 2
- 230000004075 alteration Effects 0.000 description 1
- 238000004458 analytical method Methods 0.000 description 1
- 230000008859 change Effects 0.000 description 1
- 230000007547 defect Effects 0.000 description 1
- 238000005516 engineering process Methods 0.000 description 1
- 230000007613 environmental effect Effects 0.000 description 1
- 238000002474 experimental method Methods 0.000 description 1
- 230000006872 improvement Effects 0.000 description 1
- 230000004048 modification Effects 0.000 description 1
- 238000012986 modification Methods 0.000 description 1
- 230000001537 neural effect Effects 0.000 description 1
- 238000006467 substitution reaction Methods 0.000 description 1
Images
Classifications
-
- G—PHYSICS
- G05—CONTROLLING; REGULATING
- G05D—SYSTEMS FOR CONTROLLING OR REGULATING NON-ELECTRIC VARIABLES
- G05D1/00—Control of position, course, altitude or attitude of land, water, air or space vehicles, e.g. using automatic pilots
- G05D1/02—Control of position or course in two dimensions
- G05D1/021—Control of position or course in two dimensions specially adapted to land vehicles
- G05D1/0212—Control of position or course in two dimensions specially adapted to land vehicles with means for defining a desired trajectory
- G05D1/0221—Control of position or course in two dimensions specially adapted to land vehicles with means for defining a desired trajectory involving a learning process
Landscapes
- Engineering & Computer Science (AREA)
- Aviation & Aerospace Engineering (AREA)
- Radar, Positioning & Navigation (AREA)
- Remote Sensing (AREA)
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- Automation & Control Theory (AREA)
- Control Of Position, Course, Altitude, Or Attitude Of Moving Bodies (AREA)
- Feedback Control In General (AREA)
Abstract
The invention relates to a navigation escape method under the coverage of a complex environment area of a group of robots, and in order to solve the problem of low coverage efficiency caused by the fact that the group of robots are in deadlock when the group of robots execute the area coverage in an unknown complex environment under the limitation of communication, the invention provides a navigation escape mechanism which is applicable to the unknown environment and has smaller calculation amount so as to enable the robots to escape from the deadlock. When the robot falls into deadlock in the regional coverage process, an optimal navigation escape point of the robot is found by introducing a cellular automaton mechanism, and the robot escapes from a dead zone by improving a BRRT (bidirectional rapid expansion star) navigation algorithm to carry out path planning, so that the group robot can efficiently cover a complex environment.
Description
Technical Field
The invention belongs to the field of group robot control, and relates to a navigation getting rid of poverty under the coverage of a complex environment area of a group robot.
Background
The distributed group robot area coverage is to automatically and rapidly and efficiently explore an unknown area by utilizing a large-scale robot, so as to obtain map information which is as perfect as possible or realize a certain specific task. At present, heuristic algorithms are mostly adopted at home and abroad to solve the problem of regional coverage of group robots, compared with random coverage, the coverage efficiency of the method is greatly improved, however, the method can only be used for realizing coverage of simple environments under the conditions of limited communication and the like, and for environments with complex obstacle distribution, the robots are often in deadlock in the coverage process, namely, the robots are in a certain narrow area and cannot escape to cause a large number of repeated coverage, so that the robots cannot work normally, the coverage efficiency is low, and the robots need to escape from the deadlock by adopting corresponding navigation escape technology.
The traditional navigation methods such as RRT and BRRT can enable the robot to better escape from the deadlock through path planning. However, most of the conventional navigation methods are only suitable for scenes with known coverage environments and target point positions, and have the problems of instability and large calculation amount, so that the conventional navigation methods are difficult to be suitable for group robots with high real-time requirements and limited calculation capacity and complex and changeable actual environments.
Disclosure of Invention
Technical problem to be solved
In order to avoid the defects of the prior art, the invention provides a navigation escape method for covering a complex environment area of a group of robots, and solves the problem of low covering efficiency caused by deadlock when the group of robots in limited communication execute the area covering in an unknown complex environment.
Technical proposal
The navigation getting rid of poverty method under the coverage of the complex environment area of the group robot is characterized by comprising the following steps:
step 1: the complex environment area is rasterized into square discrete units with the same size according to the interval delta L to obtain a grid map, wherein the size of each grid is the coverage area of one robot, and the grid map is endowed to each robot;
step 2: the neuron network map is updated according to the following biostimulation neural network model
Wherein: h is a c The activity value of the neurons in the grid c is the stimulation amount; i c Is the external input to grid c; l is the number of grid c neighborhood neurons; w (w) ck Is a connection weight; A. b and D are positive constants;
by giving the following input I c ：
Obtaining that the activity value of the neurons of the uncovered grid is close to 1, the activity value of the neurons of the covered grid is close to 0, and the activity value of the neurons of the barrier grid is close to-1;
step 3: judging whether the current robot falls into a deadlock state, namely judging whether the activity values of eight grid neurons around the robot are all smaller than or equal to 0, if so, indicating that all neighbor grids around the robot are detected or are obstacles, wherein the robot falls into the deadlock state at the moment, executing the step 6, otherwise executing the step 4;
step 4: determining a next decision of the robot according to the neuron activity value, namely selecting a grid with the neuron activity value close to 1 as a target point of the robot for the next step;
step 5: judging whether the robot reaches the coverage requirement, if so, ending the coverage, otherwise, returning to the step 2;
step 6: the optimal navigation getting rid of trouble point of the robot is obtained by adopting a cellular automaton algorithm, and the following rule of a cellular state value table is as follows:
1. initializing a cell state value table: the current grid cell state value of the robot is 1, the barrier grid cell state value is-1, and the rest grid cell values are 0;
2. evolution rules of a cell state value table: when the state value of the central cell is-1, the value is unchanged, otherwise, the minimum value of the cell values which are more than or equal to 1 in 8 neighbors of the central cell is added with 1 to update the state value of the central cell;
using the cell state value table updated by the rule, wherein each value in the table is the number of steps from the trapped point of the robot, and selecting the uncovered point with the least number of steps as the navigation trapped point;
step 7: the robot after executing the step 6 carries out path navigation by adopting a BRRT navigation algorithm based on the activity value of the neural network, so that the robot escapes from the deadlock state; returning to step 5.
Advantageous effects
The invention provides a navigation escape method under the coverage of a complex environment area of a group of robots, and aims to solve the problem that the coverage efficiency is low due to the fact that the group of robots are in deadlock when the group of robots execute the regional coverage in an unknown complex environment under the limitation of communication. When the robot falls into deadlock in the regional coverage process, an optimal navigation escape point of the robot is found by introducing a cellular automaton mechanism, and the robot escapes from a dead zone by improving a BRRT (bidirectional rapid expansion star) navigation algorithm to carry out path planning, so that the group robot can efficiently cover a complex environment.
The invention improves the BRRT algorithm by utilizing the characteristic that the activity value of the grid neuron which is closer to the undetected area is larger on the traditional BRRT algorithm; in the BRRT navigation algorithm method based on the neuron network activity value, the size between the neuron activity value of the new node and the neuron activity value of the father node is compared before collision detection, if the father node is a point on a starting point tree, the next collision detection is carried out when the neuron activity value of the new node is larger than the neuron activity value of the father node, otherwise, the random point is regenerated; if the father node is a point on the getting rid of the trapping point tree, performing next collision detection when the neuron activity value of the new generation node is smaller than that of the father node, otherwise, returning to regenerate the random point; the method guides the generation of random points, reduces blind sampling, and improves path planning speed and algorithm stability.
Drawings
Fig. 1: BRRT navigation algorithm based on neuron network activity value
Fig. 2: navigation getting rid of poverty method flow chart under complex environment area coverage of group robot
Fig. 3: cell state value table update process
Fig. 4: BRRT navigation algorithm simulation contrast diagram before and after improvement
Fig. 5: grid map of complex environment
Fig. 6: robot trap dead zone diagram
Fig. 7: initial position diagram of robot
Fig. 8: robot complete coverage map
Fig. 9: simulation comparison before and after adding navigation getting rid of poverty mechanism
(a) Average visit number simulation contrast map, (b) total movement step simulation contrast map
Detailed Description
The invention will now be further described with reference to examples, figures:
the invention provides a navigation escape mechanism which is applicable to unknown environments and has smaller calculation amount so as to enable a robot to escape from deadlock. When the robot falls into deadlock in the regional coverage process, an optimal navigation escape point of the robot is found by introducing a cellular automaton mechanism, and the robot escapes from a dead zone through path planning by improving a BRRT (bidirectional rapid expansion satellite) navigation algorithm, so that the group robot can efficiently cover a complex environment, and is more suitable for an actual environment.
In the invention, the coverage degree of the robot to the unknown environment is represented by the magnitude of the environmental stimulus in the area coverage process, the magnitude of the stimulus is updated by using a biological excitation neural network model, and the dynamic change rule is represented by the following formula:
wherein h is c The activity value of the neurons in the grid c is the stimulation amount; i c Is the external input to grid c; l is the number of grid c neighborhood neurons; w (w) ck Is a connection weight; A. b and D are positive constants.
By giving the following input I c ：
It is possible to obtain an uncovered grid stimulation close to 1, a covered grid stimulation close to 0, and an obstacle grid stimulation close to-1. The robot is guided to move towards the grid with the stimulation quantity approaching 1, so that the whole environment area is covered.
When the activity values of the grid where the robot is and the surrounding 8 grids are all less than or equal to 0, namely the neighbor grids around the robot are all detected or are obstacles, the robot is shown to fall into deadlock. The invention adopts a cellular automaton algorithm to calculate the optimal navigation escape point in real time. The following rules of the cell state value table are as follows:
1. initializing a cell state value table: the current state value of the grid cell of the robot is 1, the state value of the barrier grid cell is-1, and the other grid cell values are 0, as shown in fig. 3 (a).
2. Evolution rules of a cell state value table: and when the state value of the central cell is-1, the value is unchanged, otherwise, the state value of the central cell is updated by adding 1 to the minimum value of the cell values which are greater than or equal to 1 in 8 neighbors of the central cell, as shown in the figure 3 (b).
And using the cell state value table updated by the rule, wherein each value in the table is the number of steps from the trapped point of the robot, and selecting the uncovered point with the least number of steps as the trapped point of the navigation.
In order to reduce the calculated amount in the navigation planning process, the invention adopts a BRRT navigation algorithm based on the activity value of the neural network to carry out path navigation so that the robot reaches the escaping target point. Conventional BRRT navigation algorithms plan paths by randomly sampling in the target space, but blind sampling can create many invalid nodes. To reduce the computational effort, the BRRT navigation algorithm is improved by taking advantage of the feature that the grid activity value is greater nearer to the undetected area. The flow chart is shown in fig. 1:
in the BRRT navigation algorithm method based on the neuron network activity value, the magnitude between the neuron activity value of the new node and the activity value of the father node is compared before collision detection, so that blind sampling is reduced. The method greatly simplifies the calculated amount, improves the path planning speed and the algorithm stability, and the simulation comparison chart is shown in figure 4.
The general flow chart of the navigation getting rid of poverty method under the coverage of the complex environment area of the group robot is shown in the following figure 2.
The navigation getting rid of poverty method under the complex environment area coverage of the group robot comprises the following specific steps:
(1) the complex environment area is rasterized into square discrete units with the same size according to the interval delta L, a grid map is obtained, wherein each grid is the coverage area of one robot, and the grid map is given to each robot, as shown in figure 5.
(2) The neuronal network map is updated according to equations (5) - (6):
(3) judging whether the current position falls into a deadlock state, namely whether the activity values of eight grid neurons around the robot are all smaller than or equal to 0, as shown in figure 6. If the deadlock is trapped, executing the step (6), otherwise executing the step (4).
(4) Determining a next decision of the robot based on the neuron activity value, i.e. the robot selects a grid with a neuron activity value close to 1 as the target point for the next step
(5) Judging whether the robot reaches the coverage requirement, if so, ending the coverage, otherwise, returning to the step (2).
(6) And obtaining the optimal navigation getting rid of trouble point of the robot through a cellular automaton algorithm.
(7) And (3) performing path navigation on the robot after the step (6) by adopting a BRRT navigation algorithm based on the activity value of the neural network, so that the robot escapes from the trapped point. Returning to step (5).
According to the steps, simulation analysis is carried out on the group robots, and simulation results are shown in figures 7, 8 and 9.
In this embodiment, a square area with a coverage area of 100m×100m is selected, and a grid map with a coverage area of 100m×100 is discretized, and the robot system is composed of 10 to 50 robots with the same structure, and initial positions of the robots are randomly generated in the coverage area, as shown in a green grid in fig. 7. The external input E takes 100, the attenuation rate A takes 5, the upper limit of the activity value B takes 1, and the lower limit of the activity value-D takes-1.
The method has the advantages that 50 repeated independent experiments are carried out under the same condition, the experimental results of the attached drawing 9 in the simulation effect diagram show that the navigation escape mechanism solves the problem that a large number of repeated coverage are caused by the fact that the group robots are in deadlock and cannot escape when the group robots in the communication limited environment carry out regional coverage in the unknown complex environment, so that the coverage efficiency is improved to a large extent, and the high efficiency of the method is verified by comparing the coverage simulation results with the coverage simulation results under the basic biological excitation neural network model.
Variations, modifications, substitutions, and alterations are now within the scope of this disclosure for those of ordinary skill in the art without departing from the principles and spirit of the invention.
Claims (1)
1. The navigation getting rid of poverty method under the coverage of the complex environment area of the group robot is characterized by comprising the following steps:
step 1: the complex environment area is rasterized into square discrete units with the same size according to the interval delta L to obtain a grid map, wherein the size of each grid is the coverage area of one robot, and the grid map is endowed to each robot;
step 2: the neuron network map is updated according to the following biostimulation neural network model
Wherein: h is a c The activity value of the neurons in the grid c is the stimulation amount; i c Is the external input to grid c; l is the number of grid c neighborhood neurons; w (w) ck Is a connection weight; A. b and D are positive constants;
by giving the following input I c ：
Obtaining that the activity value of the neurons of the uncovered grid is close to 1, the activity value of the neurons of the covered grid is close to 0, and the activity value of the neurons of the barrier grid is close to-1;
step 3: judging whether the current robot falls into a deadlock state, namely judging whether the activity values of eight grid neurons around the robot are all smaller than or equal to 0, if so, indicating that all neighbor grids around the robot are detected or are obstacles, wherein the robot falls into the deadlock state at the moment, executing the step 6, otherwise executing the step 4;
step 4: determining a next decision of the robot according to the neuron activity value, namely selecting a grid with the neuron activity value close to 1 as a target point of the robot for the next step;
step 5: judging whether the robot reaches the coverage requirement, if so, ending the coverage, otherwise, returning to the step 2;
step 6: the optimal navigation getting rid of trouble point of the robot is obtained by adopting a cellular automaton algorithm, and the following rule of a cellular state value table is as follows:
1. initializing a cell state value table: the current grid cell state value of the robot is 1, the barrier grid cell state value is-1, and the rest grid cell values are 0;
2. evolution rules of a cell state value table: when the state value of the central cell is-1, the value is unchanged, otherwise, the minimum value of the cell values which are more than or equal to 1 in 8 neighbors of the central cell is added with 1 to update the state value of the central cell;
using the cell state value table updated by the rule, wherein each value in the table is the number of steps from the trapped point of the robot, and selecting the uncovered point with the least number of steps as the navigation trapped point;
step 7: the robot after executing the step 6 carries out path navigation by adopting a BRRT navigation algorithm based on the activity value of the neural network, so that the robot escapes from the deadlock state; returning to step 5.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202111384047.9A CN114237226B (en) | 2021-11-10 | 2021-11-10 | Navigation escaping method under complex environment area coverage of group robot |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202111384047.9A CN114237226B (en) | 2021-11-10 | 2021-11-10 | Navigation escaping method under complex environment area coverage of group robot |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN114237226A CN114237226A (en) | 2022-03-25 |
| CN114237226B true CN114237226B (en) | 2023-06-30 |
Family
ID=80750303
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202111384047.9A Active CN114237226B (en) | 2021-11-10 | 2021-11-10 | Navigation escaping method under complex environment area coverage of group robot |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN114237226B (en) |
Citations (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| WO2016045615A1 (en) * | 2014-09-25 | 2016-03-31 | 科沃斯机器人有限公司 | Robot static path planning method |
| CN106843216A (en) * | 2017-02-15 | 2017-06-13 | 北京大学深圳研究生院 | A kind of complete traverse path planing method of biological excitation robot based on backtracking search |
-
2021
- 2021-11-10 CN CN202111384047.9A patent/CN114237226B/en active Active
Patent Citations (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| WO2016045615A1 (en) * | 2014-09-25 | 2016-03-31 | 科沃斯机器人有限公司 | Robot static path planning method |
| CN106843216A (en) * | 2017-02-15 | 2017-06-13 | 北京大学深圳研究生院 | A kind of complete traverse path planing method of biological excitation robot based on backtracking search |
Non-Patent Citations (1)
| Title |
|---|
| 基于Hopfield神经网络的路径规划研究;陆亮;许双伟;;电脑编程技巧与维护(第12期);全文 * |
Also Published As
| Publication number | Publication date |
|---|---|
| CN114237226A (en) | 2022-03-25 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN109164810B (en) | Robot self-adaptive dynamic path planning method based on ant colony-clustering algorithm | |
| CN110794842A (en) | Reinforced learning path planning algorithm based on potential field | |
| CN110989612A (en) | Robot path planning method and device based on ant colony algorithm | |
| CN109116841B (en) | Path planning smooth optimization method based on ant colony algorithm | |
| CN107295541B (en) | Wireless sensor network coverage optimization method based on virtual force and firefly algorithm | |
| CN107229287A (en) | A kind of unmanned plane global path planning method based on Genetic Ant algorithm | |
| CN105527964A (en) | Robot path planning method | |
| CN111818534B (en) | Three-dimensional optimization deployment method for layered heterogeneous wireless sensor network | |
| CN111474925B (en) | Path planning method for irregular-shape mobile robot | |
| CN113917925B (en) | Mobile robot path planning method based on improved genetic algorithm | |
| CN115297484B (en) | Sensor network coverage rate optimization method based on novel compact particle swarm algorithm | |
| Yu et al. | A novel parallel ant colony optimization algorithm for warehouse path planning | |
| Liu et al. | Robot search path planning method based on prioritized deep reinforcement learning | |
| CN114756027A (en) | Mobile robot path planning method based on improved ant colony algorithm and Bezier curve | |
| Chen et al. | Path planning of mobile robot based on an improved genetic algorithm | |
| CN109242150A (en) | A kind of electric network reliability prediction technique | |
| CN110726416A (en) | Reinforced learning path planning method based on obstacle area expansion strategy | |
| CN113467481B (en) | Path planning method based on improved Sarsa algorithm | |
| CN114237226B (en) | Navigation escaping method under complex environment area coverage of group robot | |
| Zong et al. | Whale optimization algorithm based on Levy flight and memory for static smooth path planning | |
| CN111884854B (en) | Virtual network traffic migration method based on multi-mode hybrid prediction | |
| Liu et al. | An efficient robot exploration method based on heuristics biased sampling | |
| CN106022510A (en) | Multi-particle-swarm-cooperative-evolution-based simulated optimization method for human-vehicle mixed evacuation | |
| CN113219989B (en) | Mobile robot path planning method based on improved butterfly optimization algorithm | |
| CN115759199A (en) | Multi-robot environment exploration method and system based on hierarchical graph neural network |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant |